A tube-type heat exchanger, for example of the type described in my U.S. Pat. No. 3,021,117 or the references cited therein or from copending patent application Ser. No. 246,932 filed Mar. 24 1981, now abandoned can be cleaned by forcing foam-rubber scrubbing balls through its tubes. These balls are introduced into the flow conduit upstream of the heat exchanger and are recovered from the flow conduit downstream of the heat exchanger. They are spongy and are of a diameter that is greater than the inside diameter of the heat-exchanger tubes by 1 mm to 2 mm, so that when they are forced through a heat-exchanger tube they contact it roughly enough to wipe any accumulations from it. This type of arrangement is employed in a power-plant heat exchanger which cannot be shut down for cleaning, since the scrubbing balls can be circulated while it is in operation with only a modest loss in efficiency compared to a complete shutdown.
The heat-exchanger tubes can be continuously cleaned, that is some balls can be continuously circulated through it, or the cleanings can be periodic. Either way it is necessary to monitor the sizes of the scrubbing balls. This size decreases with time, as the balls are worn down by friction with the tubes. The rate of wear is dependent on several factors such as temperature, acidity or basicity, and dirtiness of the coolant water, which factors change often for a heat exchanger cooled, for instance, by a river. Once the balls get too small, it is necessary to replace them with fresh, larger-diameter ones.
European patent application No. 9,137 treats this selection of balls to be replaced purely as a geometric problem, simply determining the ball mesh size. The balls are removed from the system, passed through a sieve of appropriate mesh size, and the balls that can slip through it are discarded and replaced with fresh balls. This system is not efficient, as the procedures are mainly manual so they can only be performed periodically at most. Such periodic cleaning is also disadvantageous in that it must be done quite frequently, to be sure that the tubes are being kept clean when the water is particularly dirty, even though such frequent cleaning is not always necessary. The cost of such periodic cleaning is considerable.
Another problem with such a sieve-based solution is that a standard sieve with holes of regular shape can often reject otherwise too small balls while passing some that are otherwise too large. The orientation of a nonround object is as much a factor determining whether it will go through the sieve or not as is its actual size.
In order to overcome this sensitivity to object shape, which apparently is not solely determinative of cleaning effect, German patent document No. 3,125,493 has proposed passing the balls through a short run of tubing of fairly small diameter. The tubing is somewhat radially expansible and provided with strain gauges so its increase in size as a ball is forced by water pressure through it can be accurately translated into an output largely insensitive to object shape. Nonetheless the results obtained are still not as good as they would seem to be, since with a set of new balls the cleaning results are measurably better, indicating that some ineffective scrubbing balls are being left in the system, and probably that some effective ones are being discarded.
Another problem with these known systems is that they typically only remove a ball when it is below a predetermined size. Typically the balls start out 2 mm oversize and the sieve is set at a little under 1.5 mm oversize so that those balls whose diameters are less than the mesh size will be sorted out. Thus all of the balls can be just barely big enough, 1.6 mm oversize for example, to escape the triage, while at the same time the overall cleaning effectiveness is relatively low. It would seem that with random introduction of fresh balls, the distribution of sizes would remain fairly random also, but this is not so because the wear the balls are subjected to varies rapidly, so there is an averaging effect. This happens when, for instance, a load of gritty water passes through the tubes, causing considerable ball wear for a short period of time.